{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:RUSQMJ52WHUMPZPUZT3YLACLWF","short_pith_number":"pith:RUSQMJ52","canonical_record":{"source":{"id":"2605.14443","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-05-14T06:38:19Z","cross_cats_sorted":["cs.LG","cs.MA"],"title_canon_sha256":"e4a37b171e149cf34fa11d843f3cb70333590f947323067315c69bf304ab40f2","abstract_canon_sha256":"2b88354b4a55a7b8145f1894d1d768d6e3109b52e37c11ce0d8bbf93b62ed17d"},"schema_version":"1.0"},"canonical_sha256":"8d250627bab1e8c7e5f4ccf785804bb14a9abc586acf3c34956be808ad920ef0","source":{"kind":"arxiv","id":"2605.14443","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.14443","created_at":"2026-05-17T23:39:06Z"},{"alias_kind":"arxiv_version","alias_value":"2605.14443v1","created_at":"2026-05-17T23:39:06Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.14443","created_at":"2026-05-17T23:39:06Z"},{"alias_kind":"pith_short_12","alias_value":"RUSQMJ52WHUM","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_16","alias_value":"RUSQMJ52WHUMPZPU","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_8","alias_value":"RUSQMJ52","created_at":"2026-05-18T12:33:37Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:RUSQMJ52WHUMPZPUZT3YLACLWF","target":"record","payload":{"canonical_record":{"source":{"id":"2605.14443","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-05-14T06:38:19Z","cross_cats_sorted":["cs.LG","cs.MA"],"title_canon_sha256":"e4a37b171e149cf34fa11d843f3cb70333590f947323067315c69bf304ab40f2","abstract_canon_sha256":"2b88354b4a55a7b8145f1894d1d768d6e3109b52e37c11ce0d8bbf93b62ed17d"},"schema_version":"1.0"},"canonical_sha256":"8d250627bab1e8c7e5f4ccf785804bb14a9abc586acf3c34956be808ad920ef0","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-17T23:39:06.992823Z","signature_b64":"REWiLMAYI90XO1mtKe2IwKwT4cZDNHlAPKcwnd1KxF8evpSLawqMj6fZ9NELbVKNsQGyST1zXrFqYsozT9OQAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"8d250627bab1e8c7e5f4ccf785804bb14a9abc586acf3c34956be808ad920ef0","last_reissued_at":"2026-05-17T23:39:06.992092Z","signature_status":"signed_v1","first_computed_at":"2026-05-17T23:39:06.992092Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2605.14443","source_version":1,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-17T23:39:06Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"T8SX6t6fMndVxsZl9RVdwE6aBJp2sQyoGr9wQ3ebK8sUfYVdgx02502YVCKpRPNAlDwdjddLnfolye6ip3fGDQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-04T05:01:55.706245Z"},"content_sha256":"a34f97bb105cbb347cfd917cc7fd3da29b8b29b84b8b15e9a100c36772b7f35b","schema_version":"1.0","event_id":"sha256:a34f97bb105cbb347cfd917cc7fd3da29b8b29b84b8b15e9a100c36772b7f35b"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:RUSQMJ52WHUMPZPUZT3YLACLWF","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Prompting Policies for Multi-step Reasoning and Tool-Use in Black-box LLMs with Iterative Distillation of Experience","license":"http://creativecommons.org/licenses/by/4.0/","headline":"A reinforcement learning framework trains a lightweight prompter to optimize prompts for frozen black-box LLMs, lifting reasoning accuracy from 55% to 90%.","cross_cats":["cs.LG","cs.MA"],"primary_cat":"cs.AI","authors_text":"Ambarish Jash, Ketan Todi, Krishna Sayana","submitted_at":"2026-05-14T06:38:19Z","abstract_excerpt":"The shift toward interacting with frozen, \"black-box\" Large Language Models (LLMs) has transformed prompt engineering from a heuristic exercise into a critical optimization challenge. We propose a Reinforcement Learning (RL) framework for training learned prompting policies via iterative distillation of experience. In this architecture, a lightweight prompter model is optimized to maximize task-specific rewards for a larger, frozen worker LLM. By utilizing a contrastive experience buffer that couples scalar rewards with dense textual critiques, our approach effectively amortizes iterative prom"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"We demonstrate significant gains, improving performance from 55% to 90% in logic-intensive reasoning and 74% to 91% in tool-use tasks. Furthermore, we analyze the structural evolution of prompts, demonstrating how the policy discovers specialized algorithmic heuristics.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"The lightweight prompter model can be optimized to maximize task-specific rewards for the larger frozen worker LLM using a contrastive experience buffer that couples scalar rewards with dense textual critiques.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"Iterative distillation of experience trains prompting policies that boost black-box LLM performance on reasoning and tool-use tasks from 55-74% to 90-91%.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"A reinforcement learning framework trains a lightweight prompter to optimize prompts for frozen black-box LLMs, lifting reasoning accuracy from 55% to 90%.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"5df6bdfcd2a2a0b047ee2907afe42052625dc7b9c29b137e45ffc83c207e43da"},"source":{"id":"2605.14443","kind":"arxiv","version":1},"verdict":{"id":"739ae3d7-0cff-49de-a9da-2e64412c5217","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-15T01:47:12.311030Z","strongest_claim":"We demonstrate significant gains, improving performance from 55% to 90% in logic-intensive reasoning and 74% to 91% in tool-use tasks. Furthermore, we analyze the structural evolution of prompts, demonstrating how the policy discovers specialized algorithmic heuristics.","one_line_summary":"Iterative distillation of experience trains prompting policies that boost black-box LLM performance on reasoning and tool-use tasks from 55-74% to 90-91%.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"The lightweight prompter model can be optimized to maximize task-specific rewards for the larger frozen worker LLM using a contrastive experience buffer that couples scalar rewards with dense textual critiques.","pith_extraction_headline":"A reinforcement learning framework trains a lightweight prompter to optimize prompts for frozen black-box LLMs, lifting reasoning accuracy from 55% to 90%."},"references":{"count":58,"sample":[{"doi":"","year":2023,"title":"Large Language Models as Optimizers","work_id":"cacd55e1-79db-4dd6-9f1b-794fc09d75bb","ref_index":1,"cited_arxiv_id":"2309.03409","is_internal_anchor":true},{"doi":"","year":null,"title":"Deconstruct the golden response to understand the implicit steps, logic, and knowledge it used","work_id":"1c7935ae-dc56-4efa-b6fa-8bc750d8d3ff","ref_index":2,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":null,"title":"Prioritize Logic and Structure:For analytical, reasoning, or multi-step tasks, your improve- ments should focus on formalizing a step-by-step thinking process","work_id":"88f5566b-3eb5-4b56-9203-3c8d294c80e1","ref_index":3,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":null,"title":"The new prompt should be self-contained","work_id":"d294cc51-5669-4ca9-b971-4d66ce990f4e","ref_index":4,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":null,"title":"What is the agent supposed to do?","work_id":"f5f1296f-b14d-41f5-8aed-ae0580ec591d","ref_index":5,"cited_arxiv_id":"","is_internal_anchor":false}],"resolved_work":58,"snapshot_sha256":"11251eb7df4d762d336ba98d82d32df0cfdfac2a73c765a055ae912230b13ee1","internal_anchors":1},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"verdict_id":"739ae3d7-0cff-49de-a9da-2e64412c5217"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-17T23:39:06Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"gtdB5P5/tqVnjA6p86dkFGy3akMVNBfWkJavJwUjyWvwMfhVT5QM9Kny7DArYRPfeqdEtQDvrbsLhZOqTDd4Bw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-04T05:01:55.706779Z"},"content_sha256":"46f592538dcb3ed8b52435dd089272613b57840227e7eb22c15fb3627f406955","schema_version":"1.0","event_id":"sha256:46f592538dcb3ed8b52435dd089272613b57840227e7eb22c15fb3627f406955"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/RUSQMJ52WHUMPZPUZT3YLACLWF/bundle.json","state_url":"https://pith.science/pith/RUSQMJ52WHUMPZPUZT3YLACLWF/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/RUSQMJ52WHUMPZPUZT3YLACLWF/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-06-04T05:01:55Z","links":{"resolver":"https://pith.science/pith/RUSQMJ52WHUMPZPUZT3YLACLWF","bundle":"https://pith.science/pith/RUSQMJ52WHUMPZPUZT3YLACLWF/bundle.json","state":"https://pith.science/pith/RUSQMJ52WHUMPZPUZT3YLACLWF/state.json","well_known_bundle":"https://pith.science/.well-known/pith/RUSQMJ52WHUMPZPUZT3YLACLWF/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:RUSQMJ52WHUMPZPUZT3YLACLWF","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"2b88354b4a55a7b8145f1894d1d768d6e3109b52e37c11ce0d8bbf93b62ed17d","cross_cats_sorted":["cs.LG","cs.MA"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-05-14T06:38:19Z","title_canon_sha256":"e4a37b171e149cf34fa11d843f3cb70333590f947323067315c69bf304ab40f2"},"schema_version":"1.0","source":{"id":"2605.14443","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.14443","created_at":"2026-05-17T23:39:06Z"},{"alias_kind":"arxiv_version","alias_value":"2605.14443v1","created_at":"2026-05-17T23:39:06Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.14443","created_at":"2026-05-17T23:39:06Z"},{"alias_kind":"pith_short_12","alias_value":"RUSQMJ52WHUM","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_16","alias_value":"RUSQMJ52WHUMPZPU","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_8","alias_value":"RUSQMJ52","created_at":"2026-05-18T12:33:37Z"}],"graph_snapshots":[{"event_id":"sha256:46f592538dcb3ed8b52435dd089272613b57840227e7eb22c15fb3627f406955","target":"graph","created_at":"2026-05-17T23:39:06Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":4,"items":[{"attestation":"unclaimed","claim_id":"C1","kind":"strongest_claim","source":"verdict.strongest_claim","status":"machine_extracted","text":"We demonstrate significant gains, improving performance from 55% to 90% in logic-intensive reasoning and 74% to 91% in tool-use tasks. Furthermore, we analyze the structural evolution of prompts, demonstrating how the policy discovers specialized algorithmic heuristics."},{"attestation":"unclaimed","claim_id":"C2","kind":"weakest_assumption","source":"verdict.weakest_assumption","status":"machine_extracted","text":"The lightweight prompter model can be optimized to maximize task-specific rewards for the larger frozen worker LLM using a contrastive experience buffer that couples scalar rewards with dense textual critiques."},{"attestation":"unclaimed","claim_id":"C3","kind":"one_line_summary","source":"verdict.one_line_summary","status":"machine_extracted","text":"Iterative distillation of experience trains prompting policies that boost black-box LLM performance on reasoning and tool-use tasks from 55-74% to 90-91%."},{"attestation":"unclaimed","claim_id":"C4","kind":"headline","source":"verdict.pith_extraction.headline","status":"machine_extracted","text":"A reinforcement learning framework trains a lightweight prompter to optimize prompts for frozen black-box LLMs, lifting reasoning accuracy from 55% to 90%."}],"snapshot_sha256":"5df6bdfcd2a2a0b047ee2907afe42052625dc7b9c29b137e45ffc83c207e43da"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"paper":{"abstract_excerpt":"The shift toward interacting with frozen, \"black-box\" Large Language Models (LLMs) has transformed prompt engineering from a heuristic exercise into a critical optimization challenge. We propose a Reinforcement Learning (RL) framework for training learned prompting policies via iterative distillation of experience. In this architecture, a lightweight prompter model is optimized to maximize task-specific rewards for a larger, frozen worker LLM. By utilizing a contrastive experience buffer that couples scalar rewards with dense textual critiques, our approach effectively amortizes iterative prom","authors_text":"Ambarish Jash, Ketan Todi, Krishna Sayana","cross_cats":["cs.LG","cs.MA"],"headline":"A reinforcement learning framework trains a lightweight prompter to optimize prompts for frozen black-box LLMs, lifting reasoning accuracy from 55% to 90%.","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-05-14T06:38:19Z","title":"Prompting Policies for Multi-step Reasoning and Tool-Use in Black-box LLMs with Iterative Distillation of Experience"},"references":{"count":58,"internal_anchors":1,"resolved_work":58,"sample":[{"cited_arxiv_id":"2309.03409","doi":"","is_internal_anchor":true,"ref_index":1,"title":"Large Language Models as Optimizers","work_id":"cacd55e1-79db-4dd6-9f1b-794fc09d75bb","year":2023},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":2,"title":"Deconstruct the golden response to understand the implicit steps, logic, and knowledge it used","work_id":"1c7935ae-dc56-4efa-b6fa-8bc750d8d3ff","year":null},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":3,"title":"Prioritize Logic and Structure:For analytical, reasoning, or multi-step tasks, your improve- ments should focus on formalizing a step-by-step thinking process","work_id":"88f5566b-3eb5-4b56-9203-3c8d294c80e1","year":null},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":4,"title":"The new prompt should be self-contained","work_id":"d294cc51-5669-4ca9-b971-4d66ce990f4e","year":null},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":5,"title":"What is the agent supposed to do?","work_id":"f5f1296f-b14d-41f5-8aed-ae0580ec591d","year":null}],"snapshot_sha256":"11251eb7df4d762d336ba98d82d32df0cfdfac2a73c765a055ae912230b13ee1"},"source":{"id":"2605.14443","kind":"arxiv","version":1},"verdict":{"created_at":"2026-05-15T01:47:12.311030Z","id":"739ae3d7-0cff-49de-a9da-2e64412c5217","model_set":{"reader":"grok-4.3"},"one_line_summary":"Iterative distillation of experience trains prompting policies that boost black-box LLM performance on reasoning and tool-use tasks from 55-74% to 90-91%.","pipeline_version":"pith-pipeline@v0.9.0","pith_extraction_headline":"A reinforcement learning framework trains a lightweight prompter to optimize prompts for frozen black-box LLMs, lifting reasoning accuracy from 55% to 90%.","strongest_claim":"We demonstrate significant gains, improving performance from 55% to 90% in logic-intensive reasoning and 74% to 91% in tool-use tasks. Furthermore, we analyze the structural evolution of prompts, demonstrating how the policy discovers specialized algorithmic heuristics.","weakest_assumption":"The lightweight prompter model can be optimized to maximize task-specific rewards for the larger frozen worker LLM using a contrastive experience buffer that couples scalar rewards with dense textual critiques."}},"verdict_id":"739ae3d7-0cff-49de-a9da-2e64412c5217"}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:a34f97bb105cbb347cfd917cc7fd3da29b8b29b84b8b15e9a100c36772b7f35b","target":"record","created_at":"2026-05-17T23:39:06Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"2b88354b4a55a7b8145f1894d1d768d6e3109b52e37c11ce0d8bbf93b62ed17d","cross_cats_sorted":["cs.LG","cs.MA"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-05-14T06:38:19Z","title_canon_sha256":"e4a37b171e149cf34fa11d843f3cb70333590f947323067315c69bf304ab40f2"},"schema_version":"1.0","source":{"id":"2605.14443","kind":"arxiv","version":1}},"canonical_sha256":"8d250627bab1e8c7e5f4ccf785804bb14a9abc586acf3c34956be808ad920ef0","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"8d250627bab1e8c7e5f4ccf785804bb14a9abc586acf3c34956be808ad920ef0","first_computed_at":"2026-05-17T23:39:06.992092Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-17T23:39:06.992092Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"REWiLMAYI90XO1mtKe2IwKwT4cZDNHlAPKcwnd1KxF8evpSLawqMj6fZ9NELbVKNsQGyST1zXrFqYsozT9OQAg==","signature_status":"signed_v1","signed_at":"2026-05-17T23:39:06.992823Z","signed_message":"canonical_sha256_bytes"},"source_id":"2605.14443","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:a34f97bb105cbb347cfd917cc7fd3da29b8b29b84b8b15e9a100c36772b7f35b","sha256:46f592538dcb3ed8b52435dd089272613b57840227e7eb22c15fb3627f406955"],"state_sha256":"4a0aa2986704b1fcbe1e5d255ce63e687ef5fd1216896cd1ed78e1794f44f452"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"t74tid4QRLjoTIKcAxyCAGjXnhAN+x9C85JriT5lqMFLmmRq4TUX+Xwn4S6HoOeVL+BvoVQ70Oll/uzSRMPPAA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-06-04T05:01:55.709471Z","bundle_sha256":"5976ad07f275cfedf8b9c9c3dfbf970a6bc502e9ec2590a88dc65a879a3f61a6"}}